Verboo logo

Verboo

Verboo Code is an AI coding agent that runs in your terminal, powered by a dedicated GPU and an OpenAI-compatible endpoint. It offers flat monthly plans priced per developer, a free trial with no charge at signup, nine models you can switch mid-conversation, native and third-party CLI integrations, and a prepaid usage-based API. Runs on Brazilian servers.

Visit Website

What is Verboo?

Verboo Code is a terminal-first AI coding agent from Verboo Tecnologia, a Brazilian company that signs its page as made with intelligence in Brazil. You install it globally with a single npm command, sign in to your Verboo account, and it becomes a coding agent running inside your own terminal. Under the hood every request travels from your machine to an OpenAI-compatible endpoint and then to a GPU allocated to your plan, so inference happens on dedicated capacity rather than shared infrastructure. The company says your code never reaches OpenAI or Anthropic, that requests run on Brazilian servers, and that models are not trained on your code or your calls. The product is built around a simple argument: token caps and surprise overages break a developer's momentum mid-session, so Verboo Code charges a flat monthly price per developer instead. Four personal plans are listed on the site, Lite at twenty dollars a month with a limited weekly allowance that resets every seven days, and Pro, Max and Ultra at sixty-nine, one hundred and twenty-nine and two hundred and sixty-nine dollars a month, each unlocking a progressively larger slice of the platform's nine models. Every plan card shows the requests per minute, the concurrent request allowance and the tokens per second of its included models, plus the monthly free-trial tokens and no charge at signup caveat. Nine models are served on the platform, from fast flash variants to flagships with a one million token context window, and you can swap between the ones your plan unlocks with a model command mid-conversation without losing the thread. Compatibility is a headline feature: any CLI, editor or agent that speaks the OpenAI API can be pointed at the Verboo base URL with an API key, and the site documents integrations for OpenCode, OpenClaw, Hermes, Pi and Cursor on top of the native agent. Alongside subscriptions the company sells a prepaid usage-based API with no monthly fee and input from twelve cents per million tokens, and an Enterprise offer with a fully dedicated GPU cluster, mutual TLS and contractual privacy. A referral programme pays credits for eligible referred payments, a Discord and WhatsApp community are promoted, and the site publishes comparison pages against Cursor, Copilot, Claude Code, Codex, Devin, Aider, Cline, Windsurf, OpenCode and Antigravity. Documentation, a status page and per-call metrics in the dashboard round out the offering. The company also runs a referral programme that pays traceable credits for eligible payments from invited developers, and a partnership track with tech creators offering shoutouts, coupons and exclusive content. Pricing is quoted in your regional currency, with Pix or card accepted where available, and every plan card states its requests per minute, concurrency ceiling and model throughput up front so you can compare before subscribing. A status page, a blog, plugin listings and a documentation portal at code.verboo.ai support the product after signup.

Verboo AI Code Assistant product interface screenshot

Verboo Core Features

✦

Terminal-native agent

install with npm install -g @verboo/code, sign in to your account and start coding inside your own terminal.

✦

Dedicated GPU per plan

requests go from your machine through an OpenAI-compatible endpoint to a GPU allocated to your plan, not shared capacity.

✦

Flat monthly pricing

Lite, Pro, Max and Ultra are fixed regional prices per developer with no additional per-token charge on subscriptions.

✦

Free trial with no charge at signup

save a card, get 10 million tokens on the plan you choose, and activate only when you decide to continue.

✦

Nine switchable models

fast flash models and flagships with up to 1M context, swappable mid-conversation with /model without losing the thread.

✦

Works with any CLI

connect OpenCode, OpenClaw, Hermes, Pi, Cursor, VS Code or JetBrains with your key, the API URL and a model from your plan.

✦

OpenAI-compatible API

prepaid balance, no monthly fee, per-call metrics and zero retention on supported models, with input from $0.12 per 1M tokens.

✦

Privacy by design

code stays on Verboo servers and never touches OpenAI or Anthropic, and no models are trained on your code or your calls.

Who is Verboo for?

Verboo Code targets individual developers and small technical teams who spend most of the day in a terminal or an IDE and want one predictable monthly bill instead of metered token anxiety. It fits people who are comfortable with a CLI, who already have a workflow they like, and who would rather plug an OpenAI-compatible provider into the client they use than adopt a brand new editor. That includes freelancers, indie hackers, backend and full-stack engineers, data and platform engineers, and startup developers shipping every day. Because you can connect OpenCode, OpenClaw, Hermes, Pi, Cursor, VS Code or JetBrains, the product suits developers who are attached to their tooling but unhappy with the rate limits and unpredictable overages they hit elsewhere. It also appeals to engineers who care about knowing exactly which model they are calling, where inference runs and which limits apply, since the site makes model availability, throughput and concurrency explicit on every plan card. Budget-conscious solo developers can start on Lite, which pairs a lower monthly price with a limited weekly allowance, while heavy daily users who keep several agents running at once will get more out of Pro, Max or Ultra and their larger model sets. Teams that need total isolation, contractual privacy and a dedicated GPU cluster are pointed toward Enterprise with mTLS, and product teams that want to embed the models into their own pipelines are pointed toward the prepaid usage-based API with no monthly fee. It is not aimed at non-technical users, at people who only want a web chat assistant, or at anyone looking for a free forever coding tool: the free trial is a 10 million token balance rather than a permanent plan, each subscription covers exactly one developer, and shared or resold access is explicitly not allowed. Developers evaluating alternatives such as Cursor, GitHub Copilot, Claude Code or Devin are the natural audience, since the site publishes head-to-head comparison pages for each of them. Anyone migrating away from per-token metering will recognise the target profile: developers burned by hitting a cap in the middle of a refactor who want the monthly price to stay exactly where it was quoted. The referral programme and the active community around the product also suggest it is aimed at developers who like to recommend their tools to peers, and the comparison pages make it easy for a buyer already considering Cursor, Copilot, Claude Code or Devin to see how Verboo positions itself.

Verboo Use Cases

Refactor a large legacy component using a 1M-context model inside your terminal.

Keep Cursor, VS Code or JetBrains and swap in Verboo's base URL and API key.

Run several agent terminals in parallel within the Lite plan's eight concurrent requests.

Prototype a product feature on the prepaid API without committing to a subscription.

Give a startup squad a dedicated GPU cluster with mTLS and contractual privacy.

Try the agent risk-free by spending the 10 million free trial tokens.

Earn credits by inviting developer friends and requesting a Pix payout.

Compare Verboo Code against Cursor, Copilot or Claude Code before subscribing.

Verboo Pros and Cons

Pros

  • Flat monthly price per developer with no per-token surcharge on subscriptions.
  • Free trial grants 10 million tokens with no charge at signup and a balance that never expires.
  • Nine models with up to 1M context, switchable mid-conversation with the /model command.
  • OpenAI-compatible endpoint lets you keep your current CLI, agent or IDE client.
  • Code stays on dedicated GPUs on Brazilian servers and is never used to train models.

Cons

  • Lite is not unlimited: its weekly allowance pauses new requests until it resets.
  • Subscription plans carry speed and concurrency limits, so heavy parallel use has to wait.
  • Each subscription covers one developer, access cannot be shared or resold, and Enterprise capacity is only available through sales.

FAQ About Verboo

Verboo Pricing

FreemiumFrom USD 0.00

Flat monthly subscriptions priced per developer, billed in your regional currency with Pix or card where available: Lite $20/month with a limited weekly allowance, Pro $69/month, Max $129/month and Ultra $269/month. Every personal plan advertises 10 million free tokens with no charge at signup, and each unlocks a different set of the platform's nine models with its own speed and concurrency limits. Enterprise offers a 100% dedicated GPU cluster with mTLS and contractual privacy through sales. A usage-based API uses a prepaid balance with no monthly fee, with input from $0.12 per 1M tokens.

Check official pricing

Enterprise

Contact salescustom

100% dedicated GPU cluster, mTLS, no rate limits and contractual privacy for teams that need total isolation; available through sales rather than self-serve.

Verboo Alternatives